Spontaneous Ganglioneuroma Possibly Originating from the Trigeminal Ganglion in a B6C3F1 Mouse

Abstract

In a carcinogenicity study, a neuronal tumor in the cranial cavity was observed in a 110-week-old female B6C3F1 mouse. At necropsy, the tumor was seen at the site of the pituitary gland. Histologically, the tumor consisted of well-differentiated ganglion cells, nerve fiber/neuropil-like elements and ganglion-like cells. The tumor was composed mainly of ganglion-like cells, which were arranged in solid sheets interspersed with thin fibrovascular stroma. Nissl substance was detected at the margin in the cytoplasm of well-differentiated ganglion cells, and nerve fibers were identified by the Kluever-Barrera method. Immunohistochemically, the well-differentiated ganglion cells were positive for S-100, neurofilament protein (NF), neuron-specific enolase (NSE), synaptophysin, and chromogranin A. The nerve fiber/neuropil-like elements were positive for S-100, NF, NSE, and glial fibrillary acidic protein (GFAP), and the ganglion-like cells were strongly positive only for NSE and synaptophysin. On the other hand, there were no pituitary cells, such as prolactin-positive or adrenocorticotropic hormone (ACTH)-positive cells in the tumor tissue. Detailed histopathological examination suggested that the tumor might be a ganglioneuroma arising from the trigeminal ganglion. This report provides additional histopathological evidence of peripheral nerve neoplasms in mice.

Case Report

Ganglioneuromas are rare tumors of the peripheral nervous system in domestic animals (Koestner and Higgins 2002). Furthermore, spontaneous ganglioneuromas consisting of mature ganglion cells and neurofibrillary matrix, or ganglion cells in the complex pheochromocytomas, are rarely seen in the adrenal medulla of rats and mice (Capen et al. 2001; Hamlin and Banas 1990; Nyska and Maronpot 1999). In rats, neuronal tumors containing ganglion cells occur exceptionally in the sellar region: a few cases of ganglioneuroma (Heath 1996; MacKenzie and Boorman 1990), one case of gangliocytoma with immature neuronal cell elements (Okazaki et al. 1997), and mixed pituitary adenoma-gangliocytoma (Pace and Perentes 2001) have been reported in the pituitary gland. In addition, only one case of ganglioneuroma derived from the cranial ganglion has been reported in a rat (Fitzgerald et al. 1974). In hamsters, three cases of ganglioneuromas originating from the sympathetic ganglia have been published (Rustia and Cardesa 1975). To the best of our knowledge, however, there have been no reports of ganglioneuromas arising from any organs or tissues other than the adrenal glands in mice. Herein, we report the first case of a spontaneous neuronal tumor containing mature ganglion cells, ganglion-like round cells, and neurofibrillary matrix arising from the cranial cavity in an aging mouse.

The animal was a 110-week-old female Crj:B6C3F1 (C57BL/6N × C3H/HeN) mouse in the low-dose group of a carcinogenicity study. It was assumed that despite being found in a treated mouse, the development of this tumor was not related to the compound being studied because of the unique occurrence of this tumor among the treated animals; none of the other treated animals in this study showed any treatment-related neurological or endocrine disturbances. The animal was housed individually in an aluminum cage with a stainless steel wire mesh front and floor under barrier conditions of 23°C ± 3°C room temperature, 55% ± 20% relative humidity, and a twelve-hour light-dark cycle. The animal had free access to a standard radiation-sterilized laboratory diet (CRF-1, Oriental Yeast Co. Ltd., Tokyo, Japan) and tap water via an automatic water supply system. All animal care and procedures were performed in accordance with the Guidance for Care and Use of Laboratory Animals at the Biosafety Research Center, Foods Drugs and Pesticides. The animal was observed for clinical signs twice a day (in the morning and afternoon). The animal showed no clinical signs until terminal necropsy.

At necropsy, the nodule (4 × 3 × 3 mm) was seen at the site of the pituitary gland and was recorded as enlarged pituitary gland. After gross postmortem examinations, systemic organs including this nodule were fixed in 10% neutral-buffered formalin, embedded in paraffin, sectioned at 3 μm, and stained with hematoxylin and eosin (H&E) for routine histopathological examination. Then, serial sections of the nodule were stained by Kluever-Barrera stain, aldehyde-thionin-PAS-orange G stain and immunohistochemically with the antibodies for S-100 (dilution 1:200; Dako, Denmark), neurofilament protein (NF, clone: 2F11; Dako, dilution 1:50), glial fibrillary acidic protein (GFAP, Dako, dilution 1:100), neuron-specific enolase (NSE, clone: BBS/NC/VI-H14, Dako, dilution 1:200), synaptophysin (clone: SVP-38, dilution 1:100; CHEMICON International, Inc., Temecula, CA, USA), chromogranin A (dilution 1:400; Dako), prolactin (dilution 1:10; R&D Systems, Inc., Minneapolis, MN, USA), adrenocorticotropic hormone (ACTH, clone: 02A3, dilution 1:50; Dako), Mac-2 (clone: M3/38, dilution 1:100; Cedarlane, Burlington, Ontario, Canada), and proliferating cell nuclear antigen (PCNA, clone: PC10, dilution 1:50; Dako), using a labeled streptavidin-biotin method (Dako LSAB2 System-HRP, Carpinteria, CA, USA). Antigen retrieval for immunohistochemical staining was performed using a pressure cooker (DC2002 Decloaking Chamber, Biocare Medical, Concord, CA, USA) at 125°C for five minutes and at 90°C for ten seconds in citrate buffer (pH 6.0). Tissue sections of the normal pituitary gland and liver in the aging female mice were subjected to concurrent staining, as positive controls, with aldehyde-thionin-PAS-orange G stain and immunostaining for prolactin and Mac-2 (a macrophage/histiocyte marker of mice).

Histologically, the nodule did not exhibit features consistent with pituitary tumors, such as adenomas of the pars distalis or intermedia. The nodule was located above the sphenoid bone and well demarcated (Figure 1). Thick nerve fascicles, considered to belong to the trigeminal nerve, were found to be connected with the nodule. However, we were unable to examine the pituitary gland, since it was not observed in any of the histopathological specimens available in this animal. The nodule consisted of well-differentiated ganglion cells, nerve fiber/neuropil-like elements containing Schwann-like cells, and ganglion-like round cells (Figure 2). The well-differentiated ganglion cells and nerve fiber/neuropil-like elements were located mainly at the edge of the nodule. The nodule consisted mainly of ganglion-like round cells arranged in solid sheets, interspersed with thin fibrovascular stroma. The ganglion-like cells had round or oval nuclei and eosinophilic or amphophilic cytoplasm, with areas showing pale eosinophilic staining. In addition, these cells sporadically showed vacuolar changes in the cytoplasm. The vacuoles often became large, replaced the cytoplasm, and compressed the nuclei (Figure 3).

These vacuolated cells showed negative immunoreactivity for Mac-2, suggesting that the vacuolated cells were not macrophages/histiocytes, but degenerated ganglion-like cells. Mitotic figures were sporadically observed, and large nuclei with prominent nucleoli were also seen in the nodule. However, PCNA staining revealed a low proliferative activity of the nodule (data not shown). Nissl substance was detected at the margin in the cytoplasm of the well-differentiated ganglion cells, and nerve fibers were identified by the Kluever-Barrera method (Figure 4). On the other hand, no pituitary cells were identified in the nodule by aldehyde-thionin-PAS-orange G staining (data not shown).

Immunohistochemically, the well-differentiated ganglion cells were positive for S-100 (Figure 5), neurofilament protein (NF) (Figure 6), neuron-specific enolase (NSE) (Figure 7), synaptophysin (Figure 8), and chromogranin A, and nerve fiber/neuropil-like elements were positive for S-100, NF, NSE, and glial fibrillary acidic protein (GFAP) (Figure 9). Glial fibrillary acidic protein-positive spindle-shaped cells (Figure 9) were also seen sporadically among the ganglion or ganglion-like cells and were considered to possibly be satellite cells surrounding the ganglion cells. The ganglion-like cells were strongly positive only for NSE and synaptophysin. Antibodies against NSE and synaptophysin are considered to be helpful for the diagnosis of neuroblastic tumors such as ganglioneuromas, ganglioneuroblastomas, or neuroblastomas (Hachitanda et al. 1989). Our results indicate that the ganglion-like cells might have possibly arisen from the neuronal cells. The difference in the immunoreactivity for S-100, NF, and chromogranin A between the well-differentiated ganglion cells and ganglion-like cells might be related to the degree of cytodifferentiation. On the other hand, there were no prolactin- or ACTH-positive cells in the nodule or the surrounding tissues.

Detailed histopathological examination revealed findings highly suggestive of ganglioneuroma arising from the cranial cavity, although the lack of availability of histological sections of the pituitary gland in this animal weakened the conclusion regarding the tumor origin. As in rodents, gangliocytomas of the sellar region are extremely rare in humans. In humans, forty-three tumors have been reported, and in 65% of the cases, the sellar gangliocytoma was associated with a pituitary adenoma, and in 74%, the tumor was associated with oversecretion of pituitary hormones (Puchner et al. 1995). In addition, these tumors have been reported to be associated with acromegaly, precocious puberty, Cushing’s disease, and amenorrhea-galactorrhea (Asa 1997). Although the precise histogenesis of these tumors remains unknown, it is highly likely that these tumors have a true hypothalamic origin (Asa 1997). In the present animal, there were no clinical signs until the terminal necropsy, pituitary cells in the tumor tissue, or histological abnormalities such as lobular hyperplasia of the mammary gland or adrenocortical hyperplasia related to oversecretion of the pituitary hormones.

Although the incidence of neoplasms of the pars distalis has been reported to be approximately 15% in female B6C3F1 mice (Mahler and Elwell 1999), intrasellar gangliocytomas or mixed pituitary adenoma-gangliocytomas have never been reported. Considering this information and our examination, the present tumor was unlikely to originate from the pituitary gland. We found GFAP-positive neuropil-like elements resembling cells of the neurohypophysis located at the edge of the tumor mass. Schwann cells, satellite cells, and glial cells of ganglioneuromas have been demonstrated to express GFAP in human and animal cases (Pace et al. 2002; Porter et al. 2007; Scheithauer et al. 1997). Glial fibrillary acidic protein-positive neuropil-like elements in the present case may have represented one of the components of the tumor tissue rather than of the neurohypophysis.

The other possible primary tumor site in this animal was the intracranial ganglia, such as the trigeminal ganglion. Ganglio-neuromas arising from the intracranial ganglia are extremely rare in both animals and humans. Trigeminal ganglioneuromas have been reported in one dog (Beezley 1969) and one human (Abe et al. 1999), whereas there have been no reports of trigeminal ganglioneuromas in mice. In the present case, residual normal thick-nerve fascicles considered to belong to the trigeminal nerve were connected to the tumor mass, and this histological finding was considered as supportive evidence for a tumor originating from a ganglion.

Based on the results of the histopathological examinations described above and previous reports, we made a diagnosis of spontaneous ganglioneuroma possibly originating from the trigeminal ganglion in an aged mouse. To the best of our knowledge, this is the first report of a ganglioneuroma arising from the cranial cavity in mice, and our findings will provide additional histopathological evidence of the possibility of spontaneous ganglioneuromas arising from tissues other than the adrenal medulla in mice.